UNITED STATES PATENT AND TRADEMARK OFFICE
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`MYLAN PHARMACEUTICALS INC.,
`Petitioner
`
`V.
`
`GENENTECH, INC. AND CITY OF HOPE
`Patent Owner
`
`Case IPR2016-00710
`
`Patent 6,331,415
`
`DECLARATION OF REINER GENTZ, PH.D.
`
`Mylan v. Genentech
`Mylan V. Genentech
`IPR2016-00710
`Genentech Exhibit 2021
`Genentech Exhibit 2021
`M
`
`IPR2016-00710
`
`

`
`I, Reiner Gentz, do hereby declare and state:
`
`1.
`
`I am a citizen of Germany and reside in Belo Horizonte, Minas
`Gerais, Brazil.
`
`From approximately the end of 2003 to the present, I have acted as a
`biotechnology consultant. I am currently the President of Gentz &
`Gentz Biotechnology Consultants, which I founded around 2012, and
`which is based in Belo Horizonte, Minas Gerais, Brazil.
`
`My career in the biotechnology industry has sparmed over thirty
`years. It formally began in 1984 when, upon completion of my
`Ph.D., I was hired by Hoflinan La Roche AG in Basel, Switzerland.
`I worked there as a Senior Scientist until 1987. I then took sabbatical
`leave and served as a visiting scientist at Roche Institute of Molecular
`Biology in Nutley, New Jersey until 1989. I then returned to
`Hoffinan La Roche AG, where I worked for another 4 years.
`
`In 1993, I joined Human Genome Sciences in Rockville, Maryland,
`where I worked for the next 10 years in various executive positions.
`
`Before joining Hoffman La Roche AG in 1984, I studied at the
`University of Heidelberg in Germany, from which I received my
`Master’s degree in biology/biological sciences in 1981 and my Ph.D.
`in biology in 1984.
`
`From 1980-1984, I was a student in the laboratory of Dr. Hermarm
`Bujard, first in pursuit of my Master’s degree and then in pursuit of
`my Ph.D.
`
`I have been asked by Genentech and City of Hope to describe
`research relating to T5 promoters that I was a part of in the Bujard
`laboratory while I was a Master’s and Ph.D. student.
`
`I have also been asked to give my views on how a scientist, with a
`Ph.D. in molecular biology or a related discipline as of April 8, 1983,
`would have interpreted the work done by my group at the Bujard lab,
`as disclosed in the publication Gentz et al., Cloning and analysis of
`strong promoters is made possible by the downstream placement of a
`
`1
`
`Lug.
`
`

`
`RNA termination signal,” Proc. Natl. Acad. Sci. USA 78(8):4936-
`4940 (1981) (Ex. 2060) (“the Gentz publication”), and in U.S. Patent
`No. 4,495 ,2s0 (Ex. 1002) (“the Bujard patent”).
`
`Promoters are the regulatory elements that drive transcription of
`DNA into mRNA, the first general step in gene expression.
`Translation, the second general step in gene expression, is the process
`by which mRNA is translated into protein. By 1980, it was known
`that some TS promoters were particularly strong, i.e., they exhibited
`higher transcription levels than other promoters, but as a result, it was
`difficult to stably introduce them into plasmids in a controlled
`fashion.
`
`10.
`
`11.
`
`12.
`
`In 1980, when I began working in the Bujard lab, the group had
`already been trying to clone some of the T5 promoters into plasmids,
`but had not yet been successful in doing so.
`
`During the 1980-1981 timeframe, I collaborated with other members
`of the Bujard lab (including Dr. Bujard and Armette Langner) as well
`as several scientists from Stanford (Dr. Stanley Cohen and Annie
`Chang), to develop a way to stably introduce TS promoters into
`plasmids to facilitate further characterization of the promoters.
`
`We ultimately did this by introducing T5 promoters into a plasmid,
`designated pLBU3, that includes a strong terminator located
`downstream of the cloning site for the promoter. The pLBU3
`plasmid also includes a gene that confers tetracycline resistance
`located downstream of the terminator signal.
`
`Transformation of E. colt‘ with pLBU3 provides only nominal
`resistance to tetracycline due to the presence of the strong terminator,
`but we theorized that the insertion of strong T5 promoters should
`result in an increase in tetracycline resistance, because more RNA
`polymerases would read through the terminator and transcribe the
`gene conferring tetracycline resistance.
`
`14.
`
`Additionally, we theorized that expression levels of a gene placed
`between the T5 promoter and terminator should be high relative to
`the expression levels of the gene conferring tetracycline resistance,
`
`2
`
`it
`
`

`
`due the strength of the T5 promoter and the placement of the
`additional gene upstream of the terminator. In our experiments, we
`inserted a small fragment of the B-galactosidase gene that encodes a
`peptide of approximately 71 amino acids in between the T5 promoter
`and the terminator signal. The B-galactosidase gene fragment
`complements enzyme activity in a bacterial strain and the resulting
`enzyme cleaves the substrate, which results in a deep blue color of
`the colonies.
`
`15.
`
`16.
`
`17.
`
`18.
`
`19.
`
`Finally, we theorized that the presence of a strong terminator might
`be sufficient to balance the T5 promoter and allow for it to be cloned
`into the expression plasmid.
`
`Our experiments confirmed that the presence of the strong terminator
`does in fact balance strong T5 promoters and allow for them to be
`cloned into the expression plasmid.
`
`We also demonstrated that the presence of a balanced T5 promoter
`increased expression levels of both the [3-galactosidase gene fragment
`and the gene conferring tetracycline resistance located downstream of
`the terminator signal, with much higher expression levels of the B-
`galactosidase gene fragment being demonstrated by the presence of a
`dark blue color of the colonies on the plates.
`
`Our work was published in 1981 in Gentz er a!., Cloning and analysis
`of strong promoters is made possible by the downstream placement
`of a RNA termination signal. Proc. Natl. Acad. Sci. USA, 78(8):
`4936-4940 (1981) (Ex. 2060).
`
`After completing the work described above, during 1981-1983, when
`I was working on my Ph.D. thesis, I continued to characterize various
`T5 promoters, but also did work demonstrating that by using the
`strong promoter/terminator system in a modified plasmid, one could
`express large amounts of eukaryotic protein in E. cob‘, such that it
`constitutes 50% of the total protein in the bacteria.
`
`20.
`
`In my work, I used a murine dihydrofolate reductase (‘T)HFR”) gene
`as the eukaryotic gene to be expressed. This gene was chosen for
`further study of the strong promoter/terminator system, because it
`
`3
`
`

`
`21.
`
`22.
`
`23.
`
`encodes a relatively simple monomeric enzyme. The murine DHFR
`gene was inserted between strong T5 promoters (which included
`ribosomal binding sites) and strong terminator in the plasmid.
`
`In these experiments, a chloramphenicol acetyltransferase (“CAT”)
`gene that confers resistance to the antibiotic chloramphenicol was
`located downstream of the terminator instead of a gene conferring
`tetracycline resistance, which we had used previously.
`
`High expression levels of DHFR were obtained, while the CAT
`resistance gene was expressed at much lower levels, but at sufiicient
`levels to identify and select colonies that exhibited CAT resistance.
`
`While I was working in the Bujand lab, the lab’s research was not
`focused on methods of protein production. Rather, the lab was
`pursuing exploratory research directed to characterizing various
`promoters and understanding why some are more efficient than
`others. Dr. Buj ard’s publications from this timeframe, all of which
`report research relating to studies of promoters and transcription,
`reflect this focus. See, e.g., Kammerer, et at, Functional dissection
`of Escherichia coli promoters: information in the transcribed region is
`involved in late steps of the overall process. EMBO J. 5(l l):299S-
`3000 (1986) (Ex. 2074); Deuschle, er al., Promoters of Escherichia
`coli: a hierarchy of in vivo strength indicates alternate structures.
`EMBO J‘. 5(l l):298’7-94 (1986) (Ex. 2075); Deuschle, et aI., lac
`Repressor blocks transcribing RNA polymerase and terminates
`transcription. Proc. Natl. Acad. Sci. USA 83(l2):4l34-37 (1986)
`(Ex. 2076); Peschke, er al., Efiicient utilization of Escherichia coli
`transcriptional signals in Bacillus subtilis. J. Mol. Biol. 186(3):54'7-
`55 (1985) (Ex. 2077); Gentz & Bujard, Promoters recognized by
`Escherichia coli RNA polymerase selected by function: highly
`efficient promoters from bacteriophage T5. J. Bacteriol. l64(1):70-
`77 (1985) (Ex. 2078); Bujard, et al., Insertion of transcriptional
`elements outside the replication region can interfere with replication,
`maintenance, and stability ofCo1E1-derived plasmids. Basic Life
`Sci. 30:45-52 (1985) (Ex. 2079).
`
`24.
`
`During this timeframe, I did not contemplate using the strong
`promoter/terminator system we had created to express two different
`
`4
`
`

`
`25.
`
`26.
`
`27.
`
`28.
`
`29.
`
`eukaryotic proteins. Nor was I aware of anyone in the Bujard lab that
`used or mentioned using these plasmids in such a way.
`
`Indeed, prior to April 8, 1983, the only work I was aware of, in which
`a eukaryotic gene was expressed using our strong
`promoter/terminator system, was our lab’s work relating to
`expression of the murine DHFR gene, which as noted above, is a
`relatively simple monomeric enzyme.
`
`Therefore, I did not contemplate introducing DNA encoding both
`heavy and light chains of an antibody in between the promoter and
`terminator of the plasmids we created. Nor was I aware of anyone in
`the Bujard lab that created or mentioned creating such a construct.
`
`Furthermore, even if I had thought about trying to create such a
`construct, I would have had little confidence that I would actually be
`able to coexpress the heavy and light chain DNA and recover
`functional antibody using the plasmids we were working with.
`
`My lack of confidence would have been due in part to the fact that as
`of April 1983, the strong T5 promoters we were using were
`completely unregulated, meaning that are never repressed and
`therefore always synthesize RNA in an uncontrolled fashion.
`Controlled induction of the T5 promoter was not reported until the
`mid-1980s. As such, I do not believe that the plasmids we were
`working with would have been suitable for expression of multiple
`different eukaryotic proteins, such as antibody heavy and light
`chains.
`
`In April of 1983, there were still various uncertainties associated with
`expressing a single eukaryotic gene in a host cell. Therefore, I would
`have confined any studies using the plasmids we created in the
`Bujard lab to a single eukaryotic gene of interest and not attempted to
`introduce two diflerent eukaryotic genes into these plasmids. Indeed,
`this is exactly what I did using the murine DI-IFR gene, as described
`in paragraphs 19-22 above.
`
`30.
`
`As a result, I do not believe that a scientist with a Ph.D. in molecular
`biology or a related discipline as of April 1983, would have read the
`
`5
`
`it
`
`

`
`Gentz publication and decided to introduce heavy and light chain
`DNA into one of the disclosed plasmids. Nor I do I believe that such
`a scientist would have expected that transformation of a host cell
`with such a plasmid would result in coexpression of heavy and light
`chains resulting in production of a fimctional antibody.
`
`Indeed, in my view, expressing different eukaryotic proteins as
`separate molecules in a single host cell is a concept that is not
`contemplated by and falls outside of the scope of the work reported
`in the Gentz publication. That is because the plasmids we created
`and described contained a single promoter/terminator cassette.
`
`1 did not contemplate, nor does the Gentz publication suggest
`utilizing two promoter/terminator cassettes, which would be
`necessary to transcribe heavy and light chain DNA into separate
`mRNAs and then translate those separate mRNAs into separate
`heavy and light chain proteins.
`
`Furthermore, because we were not focused on expression of
`multimeric eukaryotic proteins in a single host cell, the Gentz
`publication contains no description at all of how one would assemble
`the constituent elements of a multimeric protein, even if one were to
`attempt to coexpress them. In my view, this would further lead a
`scientist with a Ph.D. in molecular biology or a related discipline to
`conclude that coexpression of heavy and light chains as separate
`molecules is not contemplated or suggested by the Gentz publication.
`
`The experimental work we did that is described in the Gentz
`publication is also reflected in the Bujard patent.
`
`I have reviewed the Bujard patent and understand that it includes
`additional language not found in the Gentz publication. I have been
`asked to give my views on some of this additional language.
`
`Like the Gentz publication, the Bujard patent relates to the use of
`strong promoters in combination with strong terminators. To the
`extent that a scientist with a Ph.D. in molecular biology or a related
`discipline read both the Gentz publication and the Bujard patent in
`
`31.
`
`32.
`
`33.
`
`34.
`
`35.
`
`36.
`
`

`
`37.
`
`38.
`
`39.
`
`40.
`
`41.
`
`42.
`
`April 1983, I believe that he or she would have concluded that the
`primary teachings of the two references are the same.
`
`I do not read the Bujard patent as suggesting anything more than the
`Gentz publication does about expressing multiple eukaryotic genes in
`a single host cell. As with the Gentz publication, I do not think that a
`scientist with a Ph.D. or a related discipline as of April 1983 would
`have read the Bujard patent to be suggesting expression of different
`eukaryotic genes to produce separate proteins in a single host cell.
`
`As such, I do not think such a scientist would have concluded that the
`Buj ard patent suggests coexpressing heavy and light chain genes as
`separate molecules in a single host cell, let alone that they could be
`assembled into functional antibody, even if they were expressed.
`
`Indeed, the Bujard patent does not describe expression of any protein
`of interest from the list of proteins in the patent at columns 4-5. The
`only expression product actually described was a small 71 amino acid
`fragment. Expression levels were not measured as the expressed
`fragment was used only to complement enzyme activity.
`
`As noted above, the Bujard patent does list a range of potential
`proteins of interest at columns 4-5, which includes antibodies. While
`all of the proteins theoretically could be considered for possible
`production in conjunction with a strong promoter/terminator pair,
`prior to April 1983, we had not actually produced any of the listed
`proteins rccombinantly using the system described in the Bujard
`patent.
`
`In my view, which I believe a scientist with a Ph.D. in molecular
`biology or a related discipline would have shared in April 1983, this
`list of proteins is not a specific suggestion about antibodies or the
`ways in which they could be recombinantly produced.
`
`The Bujard patent does reference including “genes” in the disclosed
`plasmids. I do not believe that any of the references in the Bujard
`patent to multiple genes, e.g., “a plurality of genes, including
`multimers or operons” or “one or more structural genes” (see Ex.
`1002, Bujard patent at col. 3, lns. 46-48 and col. 7, lns. 61-63) would
`
`7
`
`

`
`43.
`
`44.
`
`45.
`
`46.
`
`have inspired a scientist with a Ph.D. in molecular biology or a
`related discipline to attempt to coexpress heavy and light antibody
`chains as separate molecules in a single host cell as of April 1983.
`
`For example, in the phrase “a plurality of genes, including multimers
`or operons,” “multimer” is referring to genes, not proteins. In the
`context of the Bujard patent, I do not read the term “multimers” to
`mean different genes that that encode a multirneiic protein. Instead, I
`believe “1nultimers” refers to repeating units ofthe same gene, which
`theoretically could have been introduced into a plasmid to further
`boost expression levels of that gene.
`
`My belief is supported by a subsequent patent listing Dr. Bujard as an
`inventor, in which he referred to repeating units of the same DNA as
`“rnultirners.” In European Patent No. 1 532 260 B1, Dr. Bujard
`refers to a “MD sequence that may be multimerized to form e.g. a
`heptarner of MO sequences.” See Ex. 2005 at p. 3, In. 23. And this
`is specifically referred to as a “Inultimer” in claim 3 of the patent,
`which requires that “the transcription unit comprises a multimer of
`tet0 sequences.” See id. at p. 19, 111.4.
`
`While Dr. Bujard’s later patent was not filed until 2002, in the 19805,
`multiple copies of the same gene were being cloned into plasmids to
`increase stabilization and boost expression levels. See, e.g., Shen,
`Multiple joined genes prevent product degradation in Escherichia
`coli. Proc. Natl. Acad. Sci. USA 81 :4627-31 (1984) (Ex. 2070); Von
`Wilcken-Bergmann er al., A synthetic operon containing 14 bovine
`pancreatic trypsin inhibitor genes is expressed in E. colt‘. EMBO J
`5(12):3219-3225 (1986) (Ex. 2071).
`
`And other scientists during this tirnefranie, including Dr. Stanley
`Cohen and Annie Chang (the co-inventors on the Bujard patent and
`two of the co-authors on the Gentz publication) referred to repeating
`units of the same DNA as a “multimer.” See, e.g., Chang & Cohen,
`Construction and Characterization of Amplifiable Multicopy DNA
`Cloning Vehicles Derived from the PISA Cryptic Miniplasmid. J.
`Bacteriology, 134(3): 1141-1156 (1978) (Ex. 2067). This publication
`contains numerous references to multimers of plasmids, which are
`multiple copies of the same DNA. See, e.g. id. at p. 1150.
`
`8
`
`(L3.
`
`

`
`47.
`
`48.
`
`49.
`
`S0.
`
`51.
`
`As of April 1983, I would not refer to heavy and light chain DNA in
`a single plasmid as a “multimer.” Nor do I believe a scientist with a
`Ph.D. in molecular biology or a related discipline as of April 1983
`would have understood “multirners” in tile Buj ard patent to suggest
`DNA encoding a multimeric protein, such as an antibody.
`
`As discussed above, there is no suggestion in the Gentz publication
`or the Bujard patent of inserting different eukatyotic genes and an
`additional promoter/terminator cassette into the plasmids described in
`the Gentz publication and Bujard patent. An additional
`promoter/tenninator cassette would be necessary to produce different
`eukaryotic proteins, including heavy and light chains, as separate
`molecules in a single host cell.
`
`I do not attribute any special significance to the reference in the
`Bujard patent to “one or more structural genes.” See Ex. 1002,
`Bujard patent at col. 7, 1113. 61-63. This could refer to a marker gene
`and a gene of interest, multiple copies of the same gene (a multimer),
`or an operon. As noted above, at the time, I was not thinking about
`introducing DNA encoding multiple different eukaryotic proteins
`into the plasmids we were working with at all. Nor was I aware of
`anyone in the Bujard lab who conducted such an experiment or
`discussed such a construct.
`
`Furthermore, the statement that proteins can be prepared “as a single
`unit or as individual subunits .
`.
`. then joined together in appropriate
`ways,” (see Ex. 1002, Bujard patent at col. 4, lns. 19-22) does not
`convey to me a suggestion to coexpress subunits of a recombinant
`multimeric protein in a single host cell.
`
`In the context of the Bujard patent, I understand “a single unit” to
`refer to a monomeric protein. The reference to individual subunits
`contemplates a multimeric protein, but those different subunits could
`be expressed in separate host cells and joined together “in appropriate
`ways” in vitro.
`
`52.
`
`1 do not read this statement, and I do not believe a scientist with a
`Ph.D. in molecular biology or a related discipline in 1983 would have
`
`9
`
`lg.
`
`

`
`53.
`
`54.
`
`55.
`
`56.
`
`read this statement, to be a teaching that the subunits of multirneric
`proteins should or could be coexpressed in a single host cell.
`
`My understanding is consistent with the passage elsewhere in the
`Bujard patent that the “strategy described above provides a vehicle
`which can be used with one or more hosts for gene expression.” See
`Ex. 1002, Bujard patent at 8:1-3. Consistent with what we did at the
`time, one host could be used to express a single eukaryotic protein of
`interest. For example, I was working with DI-IFR.
`
`Alternatively, multiple hosts could be used if one attempted to
`express subunits of a multimeric protein. As of April 1983, the only
`recombinant eukaryotic protein of which I am aware that had been
`produced by expressing individual subunits and joining them together
`was insulin. Goeddel, er al., Expression in E. coli of chemically
`synthesized genes for human insulin, Prac. Natl. Acad. Sci. USA
`76:106-110 (1979) (Ex. 2011). And this was done by expressing the
`individual subunits, A chain and B chain, in separate host cells, or by
`expressing them as a fusion protein (a single unit) in a host cell. Id.
`
`Reading the disclosure of the Bujard patent as a whole, in April 1983
`I would not have interpreted “a vehicle which can be used with one
`or more hosts” to be a reference to coexpressing multiple eukaryotic
`genes in a single host cell. I do not think a scientist with a Ph.D. in
`molecular biology or a related discipline would have read the passage
`in this way either, in view of the lack of any specific disclosure of
`such an approach elsewhere in the Bujard patent. As noted above, I
`was not contemplating such an approach, nor was I aware of anyone
`else in the Bujard lab who was pursuing such an approach.
`
`Instead, I believe the reference to a “vehicle which can be used with
`one or more hosts,” is a suggestion that the disclosed plasmids can be
`used in different types of host cells, for example, B. subtilis and E.
`coli. Indeed, several years later, I, along with Dr. Bujard and several
`other co-authors, published a paper demonstrating that the
`transcriptional signals we were working with fimctioned in both E.
`coli and B. subtilis. See Peschke, er al., Efficient utilization of
`Escherichia coli transcriptional signals in Bacillus subtilis, 186 J.
`Mol. Biol. l86(3):547-55 (Dec. 1985) (Ex. 2077).
`
`10
`
`

`
`57.
`
`58.
`
`59.
`
`60.
`
`61.
`
`The fact that the Buj ard patent describes the use of multiple stop
`codons in each reading frame does not change my view of what it
`discloses. Like the Gentz publication, the Bujard patent discloses the
`pLBU3 plasmid, which includes a fiagment upstream of the [3-
`galactosidase gene fragment, which includes multiple stop codons in
`each of the possible translational reading frames. See Ex. 1002,
`Bujard patent at 10:9-16; see Ex. 2060, Gentz publication at 4938.
`
`The presence of these stop codons was usefiil in our work, because it
`ensured that any potential translation originating in the region
`upstream of the B-galactosidase gene fragment would be terminated
`before translation of the fragment began. Therefore, the B-
`galactosidase gene fiagment will not be expressed as a fusion protein.
`This in turn, allows for host cells transformed with plasmids
`containing the B-galactosidase gene fragment to be identified,
`because the expressed peptide complements enzyme activity in the
`host cells, which generates observable blue colonies. See Ex. 2060,
`Gentz publication at 4937, Table 1.
`
`More generally, the Bujard patent discloses that “the gene is followed
`by one or a plurality of translational stop codons .
`.
`. where there is at
`least one stop cotton in each reading frame.” See Ex. 1002, Bujard
`patent at col. 3, lns. 15-19. I do not read this passage to have
`anything to do with the question of whether multiple different
`eukaryotic genes could be inserted in a plasmid and coexpressed in
`single host cell as of in April 1983.
`
`The passage quoted above references a single gene, which will have a
`single reading frame and at least one stop eodon. Indeed, all genes
`have stop codons as part of their sequence, to ensure that translation
`of the gene is terminated.
`
`A scientist with a Ph.D. in molecular biology or a related discipline
`in April 1983 would have understood the importance of stop codons
`generally and in the context of our research. He or she would not
`have thought that discussion of stop codons in the Bujard patent or
`the Gentz publication had any specific relevance to the coexpression
`of multiple eukaryotic genes in a single host cell.
`
`11
`
`mg.
`
`

`
`I declare that all statements made herein of my own knowledge are true and that all
`statements made on information and belief are believed to be true; and further that
`these statements were made with the knowledge that willful false statements and
`the like so made are punishable by fine or imprisonment, or both, under Section
`1001 of Title 18 of the United States Code.
`
`52,014,
`Tllece/~m‘su 1?
`
`
`Date
`
`Klksudusu
`Reiner Gentz, Ph.
`
`’